Browsing by Subject "Measurement scales"

Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    Open Access
    Validating survey measurement scales for AIDS-related knowledge and stigma among construction workers in South Africa
    (BioMed Central, 2016-01-23) Bowen, Paul; Govender, Rajen; Edwards, Peter
    Background: Construction workers in South Africa are regarded as a high-risk group in the context of HIV/AIDS. HIV testing is pivotal to controlling HIV transmission and providing palliative care and AIDS-related knowledge and stigma are key issues in addressing the likelihood of testing behaviour. In exploring these issues, various studies have employed an 11-item AIDS-related knowledge scale (Kalichman and Simbayi, AIDS Care 16:572-580, 2004) and a 9-item stigma scale (Kalichman et al., AIDS Behav 9:135-143, 2005), but little evidence exists confirming the psychometric properties of these scales. Methods: Using survey data from 512 construction workers in the Western Cape, South Africa, this research examines the validity and reliability of the two scales through exploratory and confirmatory factor analysis and internal consistency tests. Results: From confirmatory factor analysis, a revised 10-item knowledge scale was developed (χ2 /df ratio = 1.675, CFI = 0.982, RMSEA = 0.038, and Hoelter (95 %) =393). A revised 8-item stigma scale was also developed (χ2 /df ratio = 1.929, CFI = 0.974, RMSEA = 0.045, and Hoelter (95 %) = 380). Both revised scales demonstrated good model fit and all factor loadings were significant (p < 0.01). Reliability analysis demonstrated excellent to good internal consistency, with alpha values of 0.80 and 0.74, respectively. Both revised scales also demonstrated satisfactory convergent and divergent validity. Limitations of the original survey from which the data was obtained include the failure to properly account for respondent selection of language for completion of the survey, use of ethnicity as a proxy for identifying the native language of participants, the limited geographical area from which the survey data was collected, and the limitations associated with the convenience sample. A limitation of the validation study was the lack of available data for a more robust examination of reliability beyond internal consistency, such as test-retest reliability. Conclusions: The revised knowledge and stigma scales offered here hold considerable promise as measures of AIDS related knowledge and stigma among South African construction workers.
  • Thumbnail Image
    Item
    Open Access
    Validation of the Schieman and Young measurement scales for work contact, work-family conflict, working conditions, psychological distress and sleep problems in construction industry professionals
    (BioMed Central, 2018-10-24) Bowen, Paul; Govender, Rajen; Edwards, Peter
    Background This study examined the construct validity and internal consistency of modified versions of the job autonomy and control, job pressure, work contact, work-family conflict, psychological distress, and sleep problems scales developed by Schieman and Young (2013) among construction professionals through confirmatory factor analysis and tests of internal consistency. Methods Using a cross-sectional design, survey data were collected from 942 South African construction professionals, of which 630 responses were considered for analysis. Confirmatory factor analysis was used to examine construct validity. Cronbach’s coefficient alpha was used to determine the internal consistency, and convergent validity was tested using correlation analysis. Results The final CFA indicated very good model fit to the data (χ2 /df ratio = 2.11, IFI = .95, CFI = .95, RMSEA = .06, and Hoelter (95%) = 176). The scales demonstrated satisfactory internal consistency: .82; .91; .83; .90; .90; and .73, respectively. Convergent validity was largely demonstrated with respect to direction of association, but not in relation to magnitude. A limitation of the validation study was the lack of available data for a more robust examination of reliability beyond internal consistency, such as test-retest. Conclusions The six scales developed by Schieman and Young (2013) hold promise as measures of work contact, work-family conflict, psychological distress, and sleep problems in relation to working conditions of construction professionals.